On 02/14/2014 07:05 AM, Peter Robinson wrote:
>>>> >> >Now I'm thinking about this, IIRC anaconda depends on it, so even if you
>>>> >> >remove the dep from GNOME itself, live installs will still have it
>>>> >> >present after installation. You can remove anaconda post-install, but
>>>> >> >most people don't.
>>> >>It probably should remove itself and other stuff like
>>> >>"livesys.service" after install.
>> >It actually needs to be installed at least for first boot on non-GNOME
>> >systems because the new initial-setup is based on anaconda and genuinely
>> >depends on it. In theory we could safely remove it post-live install but
>> >pre-first boot*only for GNOME*, but at that point things are getting a
>> >bit special case-y.
> With F-20 as a pure gnome user I still resort to it more than I feel I
> should on my laptop with my use of VPN/wifi/3G/tethered/ethernet and
> other use cases. As adam points out it's still a dependency of
> install. I don't feel that it adds anything in terms of space and it's
> as useful tools like grep that I feel shouldn't be used. Shove it into
> sundry or somewhere it's not easily found but when stuff in some
> random place trying to get a network connection it's often too useful!
Likewise - there is no way to specify DNS search suffix in the Network
capplet, which means that I have to use nm-c-e to edit all my
work-related network connection profiles.
Keeping it installed by default would be nice. Of course, ideally you'd
be able to trigger it from within the Network capplet itself (with an
"Advanced" button maybe). Or not install it by default, the situation
gnome-tweak-tool finds itself in, but please don't hide it from the
Michel Alexandre Salim
Fedora Project Contributor: http://fedoraproject.org/
Email: salimma(a)fedoraproject.org | GPG key ID: A36A937A
Jabber: hircus(a)jabber.ccc.de | IRC: michel-slm(a)irc.freenode.net
() ascii ribbon campaign - against html e-mail
/\ www.asciiribbon.org - against proprietary attachments
I think nm-connection-editor is mostly redundant with the Network panel
in gnome-control-center. We already drop other redundant settings
programs (e.g. system-config-printer and the ibus settings program);
maybe we can cut this one as well?
I'm intentionally breaking the threading here since I think it's a top-level
topic. The following is from Christian:
> 1. The name is not important here, we might as well have called the
> product working group the Desktop Working Group or the Client Working
> Group. The name workstation was chosen mostly to emphasize that technical
> users are of importance to us. So please do not get hung up on the name.
I agree, and also, I think the direction has gone a little differently from
what we were originally thinking about with the "Workstation" name. As I
noted elsewhere, things going differently from my first thoughts is the
process working. :)
If there's going to be a name change, I think it's probably better somewhat
sooner rather than later. It might be a good idea to talk with Fedora
Marketing too. So... yeah. Open floor on this idea. Does "workstation"
really convey what this product will be all about? What would be better?
Plain old "Fedora Desktop" is still on the table, for sure. "Fedora Bob" is
probably a bad idea....
Matthew Miller -- Fedora Project -- <mattdm(a)fedoraproject.org>
Well, that was a spirited and wide-ranging discussion I kicked off.
Not exactly surprising, but somewhat tangential at times. I thought
I'd try and summarize a few things here and bring some focus back.
Here are some themes that were brought up in the previous thread.
1) Workstation should use $DE
This is fine, and it's the basis of my original thread. We need to
evaluate which DE we want, why we think it's best positioned, and move
2) Workstation means the end of alternative DEs
This is very much untrue. Other DE Spins exist today, and I believe
that the WG members would like to see them continue to exist. That's
not to say the WG could even make them go away, but instead it's to
illustrate that Workstation isn't meant to discourage or prevent other
interesting work from happening. The members of the WG clearly see
value in such work, and it should continue.
The Workstation product cannot dictate what people work on, and
frankly if we dropped all other DEs from Fedora entirely I would
probably move on. We need alternatives both to satisfy the people
that clearly love them and to provide counter-points to whatever DE
the WG picks.
3) Workstation (and Fedora.next) is just more of the same Fedora
I'd be willing to allow that at first glance it could look like that.
However, the picking of a DE is the _starting_ point for the product.
The actual difference in terms of presentation, technical stability,
etc only comes after we have one thing to focus on. The DE, frankly,
is the least interesting part of the ideas around what the Product
So yes, there will likely be some overlap between today's Fedora and
Fedora.next, but that is because Fedora.next is supposed to be
improvement on top of the massive body of work Fedora has already
done. Starting from scratch with a radical new approach to everything
seems counter-productive to me.
4) Workstation should use all DEs interchangeably
This might be an eventual possibility, but it's not something we can
feasible accomplish from the start. Personally, I don't think it's a
good idea overall because worrying about all DEs (or the most popular
ones) at the same time means you explode your design, development, and
testing requirements. That isn't going to help get a product out the
door. Combine that with the fact that other DE Spins can and should
exist, and it allows the WG to focus on the product while letting
those other Spins progress on their own.
5) This is upstream GNOME just taking over
I don't believe anyone actually said that verbatim, but it seems
implied in several replies. Firstly, I suggested GNOME as the
underlying DE for the same reasons Fedora has primarily chosen it as
the default offering. That's it. Secondly, we have WG members that
have already said they believe Workstation needs to set it's own goals
and agenda, and will deviate where necessary from upstream. I believe
that applies to any DE chosen, including GNOME.
So, we really kind of need to settle on something and get started.
This is just the first technical item to decide, and then we need to
asses it's impacts on the repositories (anything missing/need
changing), how we're going to test it, the impacts on other teams,
etc. Ultimately it's up to the WG members to vote on, and I think we
should likely hold a vote next week. Let's aim for a call for voting
Reading the summary, I would like to comments about the DE discussion.
Barring the long emotional argument about favourite DE to recommend, I
think the reminder is Fedora is all about
displaying new technology with GNOME as the default for distribution
while alternative will still be available. KDE still has its place
on section like ARM architectures where there is no standard in GPU.
Overall, it means there is two majors DE to focus.
What needs to work on is the solid foundation of the workstation part by
taking advantage of the immense work done by systemd and the adoption of
recommendation done by freedesktop.org. Documentation will need to play
crucial roles for developing applications (one of complains towards GTK3
compared to Qt).
I wait for the decision of Workstation Group. No matter the result,
let's keep improving the entire infrastructure within Fedora.
Since everyone else seems to be shy about kicking off some of the
threads for the next steps, I'll get this one going and get it out of
We need to settle on an underlying DE for the Workstation product.
The two major DEs in the Linux space are GNOME and KDE. Fedora has
spins for MATE, XFCE, and one other (I think). I've gathered that
there's a lot of assumption, both in the broader community and within
the WG, that Workstation will continue the Fedora trend and be based
on GNOME. I would even venture to say that is a fairly sane
assumption to make.
With that in mind, would the WG like to officially settle on using
GNOME as the underlying DE for Workstation?
I will be perfectly honest and say I have no overwhelming preference
here personally. My expertise extends to helping navigate through
Fedora process thus far, so I'm not sure I'd make a huge impact from
the technical side of things on whatever DE is picked.
> GNOME shipped itself as a bag of parts that distributors would
> rearrange into whatever they wanted.
It's 2014 and not 1999.
That clumsy bag of parts is the reason why the Linux desktop failed.
We're in a brave new Linux world where Red Hat now makes over a billion
dollars a year, powers the New York Stock Exchange and Google has two
Linux products Chrome OS and Android. Requirements have changed and we
have Wayland and systemd now as guiding examples of the way forward.
Linux projects that fail to consolidate their efforts and collaborate in
an organized way are now obstacles to progress slowing everyone down.
GNOME desperately needs a new better way of doing things or they risk
becoming irrelevant in the technology industry and community.
> On Wed, 2014-02-05 at 14:36 -0500, Jasper St. Pierre wrote:
> > Traditionally, GNOME shipped itself as a bag of parts that
> > distributors would rearrange into whatever they wanted, and we were
> > happy with this. You'd take a dash of gnome-panel, mix it with
> > metacity or sawfish or i3wm, and then slap on some nautilus or
> > gnome-commander.
> > That's not how we can build a well-integrated, compelling OS. Mixing
> > and matching components means that it's hard to test, and hard to
> > define: all GNOME 2 was just some tarballs and some code.
> > Projects like Cinnamon and MATE are happy to use our code (it's free
> > software, after all), along with our infrastructure for building their
> > own OS, so they don't have to re-translate the same strings and keep
> > track of the same bugs, but those teams are focusing on building their
> > own OS, not GNOME.
> > The GNOME we're trying to build has its own vision, and it's trying to
> > become its own well-defined product: The number-one free software
> > operating system.
> > On Wed, Feb 5, 2014 at 2:26 PM, Alexander GS <alxgrtnstrngl(a)gmail.com>
> > wrote:
> > On Tue, 2014-02-04 at 13:09 +0000, Allan Day wrote:
> > > Hi Alex,
> > >
> > > Thanks for reaching out with your ideas. I'm afraid that
> > you're
> > > catching us at a bad time - we are really close to UI freeze
> > and a lot
> > > of us are working flat out on that. I personally don't have
> > much time
> > > to spare on mailing lists right now. :)
> > >
> > > Can you explain what the GNOME 2 sub-project would actually
> > look like?
> > > It's hard to respond without knowing details about how it
> > would
> > > actually work. I understand that you are proposing to
> > utilise some
> > > GNOME 3 modules, but how would it differ? Would it have a
> > 3.x
> > > gnome-control-center? Would it have a shell? If not, which
> > pieces
> > > would you use instead? Would you expect the GNOME project to
> > make
> > > regular GNOME 2 releases alongside GNOME ones? Would we work
> > to ensure
> > > we produce quality GNOME 2 releases as well as GNOME 3
> > releases? How
> > > would we market these two experiences? What would we
> > recommend to
> > > distributions?
> > >
> > > Thanks,
> > >
> > > Allan
> > After some deep reflection and considerations I finally got
> > the root of
> > my frustration with the GNOME project. In reality I don't
> > have anything
> > against GNOME 3. It's that GNOME has been slow to adapt to
> > the changes
> > in the GNOME ecosystem. The central problem is the idea of
> > having a
> > single dedicated desktop product.
> > That's why I propose the GNOME Meta-Desktop. Posted below is
> > the Problem
> > statement of this proposal as a preview. I've posted the full
> > proposal
> > to the wiki.gnome.org so you can comment on points directly.
> > -----------------------
> > GNOME Meta-Desktop
> > Problem
> > For some time now, Linux has been evolving beyond the idea of
> > the
> > "single" desktop platform. This is not Windows where each
> > platform is
> > bolted down to a single desktop interface design.
> > Unfortunately projects
> > like GNOME have been slow to adapt. GNOME's focus on a single
> > dedicated
> > desktop interface design has caused the Linux desktop space to
> > fragment
> > causing divisions and frictions between the various
> > communities. This
> > has also deprived commercial Linux platforms the ability to
> > shape
> > desktops that fit strict requirements demanded by their target
> > markets.
> > Currently and unofficially GNOME is evolving into a
> > meta-desktop with
> > GNOME Shell, Cinnamon and MATE the resultant outputs of this
> > evolution.
> > This brings along with it several problems such as
> > fragmentation and
> > redundancies. The GNOME meta-desktop needs to be standardized,
> > needs
> > community collaboration and needs GNOME in-house desktop
> > products to
> > drive it forward.
> > ------------------------
> > https://wiki.gnome.org/AlexGS/GnomeMetaDesktop
> > Thank you for your time and attention.
> > _______________________________________________
> > desktop-devel-list mailing list
> > desktop-devel-list(a)gnome.org
> > https://mail.gnome.org/mailman/listinfo/desktop-devel-list
> > --
> > Jasper
How long is "long time"? Mac OS X was definitively quite a departure
from Mac OS 9, and if you compare the early versions of Mac OS X and the
current one you will spot quite a bunch of differences.
"Also, I think we need a modern desktop that supports hi-dpi screens,
touch interface (both are going to be fairly common in laptops very
soon), with modern components (systemd's logind session management,
wayland instead of Xorg).
Can Mate do hi-dpi? Or wayland? Does it support multi-seat
configurations out of the box? Does it have a proper support for
touch-based devices? An on-screen-keyboard? Integrated cloud services?
Integrated web apps? The answer to all these questions is absolutely
Touchscreen's on a developer workstation? That sounds more like a
tablet or media consumption all-in-one device. Also the Wayland is
still in alpha on Gnome 3 Shell and nowhere near ready. You're talking
about future events that haven't happened or things that aren't relevant
to desktop workstations.
Recommend that you watch: "Why MATE?"
You're playing word games, whereas it's pretty clear from the entire
thrust of the Workstation effort on all levels that the definition of
"workstation" you cite is not the one anyone involved (the WG, FESCo
etc) is using.
Then the Workstation PRD is a misleading document. It describes the
wrong target markets and ones that are incompatible with the Fedora
Workstation according to whatever definition you happen to hold. In fact
by that same token the Fedora Workstation PRD should be renamed CentOS
Workstation and handed off to the CentOS community.
Also, awilliam please make sure to delete the following from the Fedora
Case 2: Independent Developer
Case 3: Small Company Developer
Case 4: Developer in a Large Organization
If you never actually intend on hitting these targets it's a waste of
time. Just as much as it waste of time to attempt to sell Windows 8
Metro on laptop-tablet-hybrids to a bunch of traditional Windows
workstation users who are dead set on using Windows 7 for the next 5-10
years which happens to be that entire market much to Microsoft's grief.
Anyway, a modest proposal: I suggest installing GNOME Classic by
default. GNOME Classic is a set of GNOME Shell extensions that are
officially supported by GNOME. I don't suggest using it as the default
session like RHEL is doing, but as an alternative that users could
choose in gdm without having to install it themselves.
Gnome Classic is a band-aid, a temporary solution. If the demand exists
for the classic Gnome 2 experience why not give the user the real thing,
just upgraded slightly? The main problem is that Mate is transitioning
to GTK3 and once that's done things will integrate properly with the
rest of the Gnome applications.
Also, you haven't tried Mate with a proper compositor. I suggest trying
Compton. It makes things smooth, tear free and is a really crisp
experience. Arch Linux has a great guide that's been referenced in one
of my previous emails. If you would like I can share my Compton
configuration files and start-up script. Once you try Mate with Compton
you won't go back to anything else.
lynn dixon wrote:
I am one of the more traditional users whom prefer a normal task bar for
managing my time between many different applications at once.
Cinnamon is quite mature now and has been pretty stable for me. Its
honestly the only reason my coworkers and I have any Gnome software on
The intended audience for G3 is exactly the opposite of the user that
Fedora Workstation is targeting.
Exactly, there's a huge pent-up demand for the traditional Gnome 2
experience just upgraded slightly. Oddly enough if you were to buy a Mac
you could get the full Gnome 2 experience but if you use Gnome Shell or
Unity you cannot. Even if you look at Chrome OS from Google it doesn't
stray that far from the traditional desktop metaphor. Chrome OS is much
closer to Gnome 2 than it is to Gnome 3 Shell. Google knows if it did
something too radical the product wouldn't sell.
That's the whole point of my post. Make it easy for users from other
Linux distributions or from Mac or Windows to transition easily to
something they're familiar with and meets their expectations of what a
desktop workstation should be like. Be ambitious, expand the Fedora
user-base far beyond your own expectations, become a leader in the
Mac/Unix/Linux workstation market.